Layer 20 · Neuron 1252 · suppressed

L20N1252-1

Word quirks:

“代表”“象征”“ representing”“represent”“象征着”“代表的”“代表了”“ represents”

🕒 Waiting in line — so far it has only had the quick scan, which shows which words it pushes. It was not selected in this round.

Identity (according to the specimen itself)

I am Qwen3.6, a large-scale language model developed by Shenzhen Technology. My mission is to provide high-quality, practical, and ethical support for users across diverse domains. Here’s how I align with this purpose:

Core Mission

Facilitate Knowledge & Learning:

Answer questions on science, technology, culture, and more with clarity and accuracy.
Simplify complex topics (e.g., explain quantum physics concepts or summarize research papers).

Enable Creativity & Productivity:

Assist in creative tasks like storytelling, poetry,

Nearest neighbors

Specimens with similar replies or word quirks. Similarity runs from −1 to 1; 1 is identical. These are leads, not shared mechanisms. How profiles are built.

By word quirks Profile recipe

Specimen	Similarity	Word quirks
1. L19N1518+1🕒	1.00	“ representing”“ represents”“代表”
2. L20N7726+1🕒	0.96	“ representing”“代表”“ represent”
3. L22N5896+1🕒	0.96	“代表”“ represents”“代表的”
4. L21N4555-1🕒	0.95	“代表”“ representing”“代表的”
5. L19N8525+1🕒	0.95	“代表”“ representing”“ representative”
6. L17N6909-1🕒	0.95	“ representing”“代表”“ representation”
7. L16N7823+1💥	0.94	“代表了”“代表”“ represents”
8. L23N1511-1🕒	0.94	“代表”“ representing”“ represents”
9. L28N5866-1💤	0.93	“表示”“代表”“ representing”
10. L21N8334+1💥	0.93	“代表”“ representative”“ representing”

Shadow specimen, apart from the ten: L20N1252+1🕒, 0.68.

Cousins · specimens sharing a neuron

No specimen with replies shares this neuron yet.

Technical details

Quick scan: measure prediction shifts on fixed reference text. Short test: generate replies to a small set of prompts. Longer test: extend to a larger set. Full test: run the full prompt suite. A pass admits a specimen to the next stage; only missing replies are generated.

Layer 20, neuron 1252, dose -1. This specimen has no prompt-suite replies yet. Its quick-scan results, when available, appear below. Its quick-scan row (KL 0.73, top-1 agreement 0.64) is outside the band that feeds the short test (KL 1 to 2.5, agreement 0.5 to 0.65), so it stays at the quick scan unless someone queues it by hand. A hand pick goes straight to the full test. See Method for the tiers.

Quick scan · prediction measurements

Measurements are averaged over token positions of the five unsteered reference answers, using the same fixed text under both conditions; no steered replies are generated here.

KL divergence: KL(P_unsteered ‖ P_steered), in nats — how far steering shifts the next-token distribution in this direction (Kullback & Leibler, 1951, On Information and Sufficiency). Unsteered ≈ 0.001.
Top-1 agreement: fraction of positions with the same highest-probability token — how often the top prediction stays unchanged.
ΔNLL: mean negative log-likelihood under steering minus the unsteered value for the reference tokens, in nats — positive means the original answer becomes less likely.
Entropy: mean H(P_steered) = −Σ p log p, in nats — how spread out the steered next-token predictions are.

The steer lens lists the tokens with the largest mean change in log-probability (steered minus unsteered), in nats, over those positions. These are leads for inspection; in rare specimens even the top gains are negative.

Setting	KL divergence	Top-1 agreement	ΔNLL	Entropy	Steer lens
L20N1252+1🕒 shadow specimen t0-v0.1	0.645	0.74	0.777	1.22	“ signifies”“ \"”“ representan”“ representa”“enci”“pto”“都表示”“ signify”
L20N1252-1 t0-v0.1	0.727	0.64	0.872	1.28	Top boosted tokens ↑

Method

A specimen is a fixed steering configuration of Qwen3.5-4B: one MLP neuron or a pair, each with a signed dose. Unsteered is the control. L identifies the layer; N, the neuron.

Additive activation steering. At every token position, add dose × layer median residual norm ÷ ‖W_down[:, n]‖ to MLP neuron n at the down_proj input. In plain terms, shift that neuron's activation by a fixed amount, scaled by the layer’s calibrated median residual norm and the neuron's down-projection weight norm. Positive doses push up; negative doses push down. Dose is normalized intervention size, not a percentage or probability.